Despite a general consensus that autism has biological causes, the specific links between brain loci of abnormality and cognitive-behavioral impairments remain unknown. Functional neuroimaging studies do not provide a clear picture, possibly due to etiological heterogeneity of autism samples. We will explore potential abnormalities of functional brain organization in autism with specific focus on individual variability, using functional MRI (fMRI) -- an imaging technique with sufficient power for individual analyses. Thirty verbal individuals with autism, 24 subjects with Asperger syndrome (AS), and age, gender, and handedness-matched controls will be studied. Our project will focus on language -- a domain which not only differentiates autism and AS in terms of delay and deficit, but is also characterized by a wide spectrum of abilities within the adult autistic population. FMRI will be performed during auditory and phonological discrimination, and lexical semantic association. Several complementary data analysis paths (groupwise and intraindividually; in normalized and native space) will enable us: (i) to identify consistent abnormalities of functional maps for autism and AS groups; (ii) to quantify activation foci in each subject in terms of spatial divergence from normal and individual variability within patient groups; (iii) to relate varying patterns of abnormal functional maps with measures of language skill; and (iv) to distinguish effects of pathological disturbance of functional maps from those of compensatory plasticity. We will test hypotheses, according to which in autism (and to a lesser extent in AS) pathological disturbance of neural differentiation and impoverished interactive language experience during critical developmental periods result in aberrant neurofunctional maps and abnormally pronounced individual variability. Language delay in autism is further hypothesized to be associated with atypical hemispheric dominance for language. Finally, we expect that individually variable patterns of abnormal functional maps will be correlated with level of language outcome. Identifying such links between brain maps and level of functioning achieved in adults will elucidate the significance of individual variation in autism in terms of different pathways of etiology and abnormal development.